The problems associated with the corrosion of reinforcing steel in concrete are now well understood. Steel reinforcing has generally performed well over the years in concrete structures such as bridge decks and parking garages, since the alkaline environment of portland cement causes the surface of the steel to "passivate" such that it does not corrode. Unfortunately, a dramatic increase in the use of road salt in the early 1960's together with an increase in coastal construction resulted in a widespread deterioration problem.
This problem developed because chloride ions, whether contained in deicing salt, in sea water, or added to fresh concrete, destroy the ability of concrete to keep the surface of the steel in a passive state. It has been determined that a chloride concentration of 0.6 to 0.8 Kg per cubic meter of concrete is the critical value above which corrosion of steel in concrete can occur. The resulting corrosion products occupy 2.5 times the volume of the original steel, and this exerts tensile stresses on the surrounding concrete. When these stresses exceed the tensile strength of the concrete, cracking and delaminations develop. With continued corrosion, freezing and thawing, and traffic load, further deterioration occurs and potholes develop.
Major research and development efforts in the field of concrete quality, construction practices, surface sealers, waterproof membranes, coated reinforcing steel, speciality concretes, and corrosion inhibitors have improved the status for new deck construction. It is generally agreed that new bridge decks constructed using selected protection systems will exhibit a long life with few maintenance problems. But many concrete structures built prior to the mid 1970's are in large part salt contaminated and continue to deteriorate at an alarming rate. Cathodic protection is recognized as the only means of stopping corrosion of steel in concrete without complete removal of the salt contaminated concrete.
Cathodic protection reduces or eliminates corrosion of a metal by making it a cathode by means of an impressed DC current or by attachment to a sacrificial anode. In this way external energy is supplied to the steel surface forcing it to function as a current receiving cathode and preventing the formation of ferrous ions. Cathodic protection was first applied to a reinforced concrete deck in June 1973. Since that time, understanding and techniques have improved, but the impressed current anodes used to distribute current to the reinforcing steel continue to be a major limitation. The anode should have the following properties:
1. Ability to withstand traffic loads and environmental conditions.
2. Design lifetime equal to or greater than the wearing surface life.
3. Sufficient surface area such that premature deterioration of the surrounding concrete does not occur, and that a good distribution of current is provided to the reinforcing steel.
4. Economically justifiable to install and maintain.
Historically, three different types of anodes have been used for cathodic protection of steel in concrete bridge decks: conductive overlays, slotted non-overlay, and distributed anodes with non-conductive overlay.
The conductive overlay was the first anode to be used and is still regarded as a useful system. In this case the anode typically consists of a mixture of asphalt, metallurgical coke breeze, and aggregate in conjunction with high silicon cast iron serving as the current contact. This system provides very uniform current distribution over the deck surface, and because the anode surface area is high, no evidence of acid or other chemical attack from anodic reaction products has been found on the underlying portland cement. The coke-asphalt overlay has exhibited structural degradation in a number of instances, however, and the time to replacement is limited to a few years. Also, freeze-thaw deterioration of improperly air-entrained concrete beneath the overlay has limited its use to decks with proper air-void systems.
Slotted non-overlay anodes were developed to extend anode life and applicability, and to realize a system which would not increase the dead load and height of the bridge deck. In this system parallel slots are first cut into the deck approximately 30-45 cm. apart. The slots are filled with a "conductive grout" mixture of carbon and organic resin which serves as the anode surface. Because the conductive grout has a limited conductivity, current is distributed to the anode by a system of platinized metal and carbon strand conductors. This anode exhibited adequate strength and freeze-thaw durability, but because its surface area is small, the adjacent concrete often experiences attack from the acid and gases which are a product of the anodic reaction. Also, distribution of current to the reinforcing steel is not ideal since the slots are widely separated. Failure was also experienced due to cracking or some other discontinuity since there is not a redundancy of current connections. Furthermore, this system is labor intensive and difficult to install.
Distributed anodes with ionically conductive overlays are similar to slotted systems, but are often easier to install. In one modification the conductive polymer grout anode is placed directly on top of the existing deck surface, together with platinized metal wire and carbon strand current conductors, and the anode is overlaid with latex-modified or conventional concrete. Rigid non-conductive overlays are often favored because they extend the deck life, retard additional salt penetration, minimize freeze-thaw damage to underlying concrete, and provide a new skid resistant riding surface. This system still experiences the same disadvantages as the slotted system regarding current distribution, acid or gas attack, and lack of redundancy.
An alternative anode for use with rigid ion-conductive overlays utilizes a flexible polymeric anode material which does not require a conductive backfill. It is produced as a continuous cable and woven into a large mesh, placed on the deck and covered with a conventional rigid overlay. This system is less time consuming to install, but still has the disadvantages of current distribution, acid or gas attack, and lack of redundancy. Such polymer anodes have been described in U.S. Pat. Nos. 4,473,450 and 4,502,929. As commercially offered, these polymer anodes are woven into a mesh with voids measuring about 20 cm. by 40 cm. Any breakage of the cable at a given point will thus impair the cathodic protection effect over a considerable area. Also the thickness of the cable (about 8 mm) is a limitation where only thin overlays are desirable.
A fourth type of system has more recently evolved for use on substructures in which the anode material is painted or sprayed directly on the concrete surface. For example, carbon loaded paints and mastics can be applied to the concrete. This provides a large anode area and ideal current distribution to the reinforcing steel. Additional platinized wire or carbon strand current connectors are needed, however, since the resistivity is high, and the anode material often peels off resulting in a short lifetime.
For example, published UK Patent Application 2 140 456A describes a conductive overlay system in which a conductive paint is applied to the surface of concrete to form an anode film. Primary anodes of platinized titanium or niobium are spaced apart each 10-50 meters for the supply of current to the anode film and thus serve essentially as current lead-ins.
An anode of flame-sprayed zinc has also been used (see for example U.S. Pat. No. 4 506 485). Originally it was thought that zinc would function as a natural galvanic anode therefore eliminating the requirement of DC power supply. It has since been established that the fixed natural voltage of zinc is too low to throw the current for sufficient distance through the concrete, however, and a power supply and current distribution system are still required. This problem coupled with the problem generated by the expansive corrosion products of zinc, have lead to minimal use of sacrificial anode systems on bridges.
With the exception of the system using zinc anodes, every system for cathodic protection of reinforcing steel in concrete has to date used carbon as the electrochemically active-anode surface. Carbon was probably first used because of its extensive use as an anode in traditional cathodic protection. It was also used because cathodic protection in concrete requires-very low current densities, which infers a very large anode surface area. This implies that the anode must be low cost, and carbon is relatively inexpensive.
Since pure carbon is not available in a structure which would be suitable for use in concrete, carbon was used as a conductive filler in organic resins, thermoplastic polymers, paints, and mastics. This technique put carbon into a physical form which could be used in conjunction with concrete, but other disadvantages of carbon remain. Carbon has a low electrical conductivity relative to metals, requiring an elaborate system of current conductors. Also, carbon is thermodynamically unstable as an anode, reacting to form carbon dioxide CO.sub.2, carbonic acid H.sub.2 CO.sub.3, and carbonates HCO.sub.3.sup.- and CO.sub.3.sup.2-, reaction products which are potentially harmful to portland cement. These reactions are known to be kinetically slow, but the effect of such reactions on anode lifetime may still be significant since, when in contact with a solid electrolyte such as concrete, even a small amount of oxidation will disrupt the anode-electrolyte interface causing a loss of electrical contact. Finally, carbon is a poor anode from the standpoint of electrochemical activity. Single electrode potentials at carbon anodes will be relatively high when operated in chloride contaminated concrete resulting in the release of chlorine gas Cl.sub.2, and hypochlorite ClO.sup.-. These reaction products will probably not be harmful to concrete, but they are strong oxidizers which react with the organic binders used, again causing concern over anode lifetime.
In summary, none of the anodes used to date exhibit all of the properties desirable for cathodic protection of steel in concrete. Although many appear to be economically justifiable, many lack sufficient area to prevent. deterioration of the concrete adjacent to the anode, many do not result in an ideal current distribution, and all present serious questions about anode lifetime. Zinc anodes are oxidized to zinc oxide which disrupts the anode-concrete interface. All anodes containing carbon operate at a high single electrode potential and generate chlorine, acid, and carbon dioxide, products which are likely to cause eventual damage to the adjacent concrete and to the organic matrix used to bind the carbon.
Electrocatalytically active anodes with valve metal substrates are known and have been successfully used in a number of applications, in particular chlorine, chlorate and hypochlorite production and as oxygen-evolving anodes in metal winning processes. Generally, the cost of such electrodes makes them particularly advantageous in "high" current density applications, e.g., 6-10 KA/m.sup.2 for chlorine production in a mercury cell or 3-5 KA/m.sup.2 in a membrane cell. Such electrodes have also been proposed for cathodic protection, but have found only limited applications in this area. In one typical cathodic protection arrangement, a wire of platinized copper-cored titanium is used to protect a metal structure. PCT Application W080/01488 described such an arrangement in which the platinized wire is wound around an insulating rope. UK Patent Application 2 000 808A proposed replacing the conventional platinized wires or rods with a channel-sectioned valve metal strip having anodically active material on the U or V-shaped spine.
Platinized valve metal meshes have also been proposed for cathodic protection of certain structures. See for example "Corrosion/79" paper number 194 which describes use of a rigid titanium expanded mesh measuring less that 0.05 m.sup.2 and coated with a layer of 1-15 micron of platinum capable of carrying a current density of 2.15 A/dm.sup.2. This was used as a discrete anode in groundbeds containing carbonaceous backfill. Rigid anode meshes of this type having an overall area up to 0.5 m.sup.2 have been offered as discrete anodes for the protection of remote structures.
U.S. Pat. No. 4,519,886 describes a linear type of anode structure for the cathodic protection of metal structures comprising a plurality of cylindrical anode segments spaced along and connected to a power supply cable. The cylindrical anode segments may be made of expanded titanium bent to shape and coated with a mixed metal oxide coating.
Obviously, none of the known coated valve metal electrodes including those proposed for other cathodic protection applications would be suitable for the cathodic protection of concrete structures. In particular, the anode designs are unsuitable for installation in this application and the cost of protecting an installation would be prohibitive.